CN107499373B

CN107499373B - Modified static tire model with no torque sensor assistance at zero to low vehicle speeds

Info

Publication number: CN107499373B
Application number: CN201710448435.6A
Authority: CN
Inventors: T·M·瓦伦吉卡尔; A·J·尚帕涅
Original assignee: Steering Solutions IP Holding Corp
Current assignee: Steering Solutions IP Holding Corp
Priority date: 2016-06-14
Filing date: 2017-06-14
Publication date: 2020-03-06
Anticipated expiration: 2037-06-14
Also published as: DE102017112968B4; CN107499373A; DE102017112968A1

Abstract

The invention provides a method of controlling an electric power steering system. The method estimates a steering rack force caused by the tire and a ground surface contacted by the tire in response to determining that one or more steering wheel torque sensors are not activated. The method generates a steering assist torque command based on the estimated steering rack force. The method controls an electric power steering system using a steering assist torque command.

Description

Modified static tire model with no torque sensor assistance at zero to low vehicle speeds

Cross Reference to Related Applications

This application is a continuation-in-part application of U.S. application No. 14/486,392 filed on 9, 15, 2014, which is incorporated herein by reference in its entirety.

Technical Field

The present invention relates to a method of controlling an electric power steering system and a power steering system.

Background

In a typical Electric Power Steering (EPS) system of a vehicle, a steering wheel torque sensor is used to determine an assist torque requested by a driver. When the steering wheel torque sensor is not activated and is not functioning properly, the EPS system may not be able to provide steering assist torque. Some methods provide for assisted miss detection for rolling vehicle speed.

Disclosure of Invention

In one embodiment of the present invention, a method of controlling an electric power steering system of a vehicle is provided. The method estimates a steering rack force caused by tires of the vehicle and a ground surface contacted by the tires in response to determining that one or more steering wheel torque sensors of the vehicle are not activated. The method generates a steering assist torque command based on the estimated steering rack force. The method controls an electric power steering system using a steering assist torque command.

In another embodiment of the present invention, a vehicle system includes a control module and a power steering system including one or more steering wheel torque sensors. The control module is configured to estimate steering rack force caused by tires of the vehicle and a ground surface contacted by the tires in response to determining that the one or more steering wheel torque sensors are not enabled. The control module is further configured to generate a steering assist torque command based on the estimated steering rack force. The control module is further configured to control the electric power steering system using the steering assist torque command.

These and other advantages and features will become more apparent from the following description taken in conjunction with the accompanying drawings.

Drawings

The subject matter which is regarded as the invention is particularly pointed out and distinctly claimed in the claims at the conclusion of the specification. The foregoing and other features and advantages of the invention are apparent from the following detailed description taken in conjunction with the accompanying drawings in which:

FIG. 1 shows a functional block diagram of a steering system including an assist torque calculation system according to an exemplary embodiment of the present invention;

FIG. 2 illustrates a data flow diagram showing an assist torque calculation system according to an exemplary embodiment of the present invention;

FIG. 3 depicts a data flow diagram of a rack load estimator in accordance with an exemplary embodiment of the present invention;

FIG. 4 illustrates a data flow diagram of an assist torque command generator according to an exemplary embodiment of the present invention;

FIG. 5 depicts a data flow diagram of a steering wheel angle based zoom module in accordance with an exemplary embodiment of the present invention;

FIG. 6 shows a flowchart of an assist torque command generation method according to an exemplary embodiment of the present invention;

FIG. 7 depicts a data flow diagram of a rack load estimator in accordance with an exemplary embodiment of the present invention;

FIG. 8 illustrates a data flow diagram including a friction estimation module according to an exemplary embodiment of the present invention;

FIG. 9 illustrates a data flow diagram including a friction learning module according to an exemplary embodiment of the present invention;

FIG. 10 illustrates a data flow diagram including a flag updater module according to an exemplary embodiment of the present invention;

FIG. 11 shows a data flow diagram including a converter module according to an exemplary embodiment of the invention; and

fig. 12 shows a data flow diagram including a mixer module according to an exemplary embodiment of the invention.

Detailed Description

The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, or uses. It should be understood that throughout the drawings, corresponding reference numerals indicate like or corresponding parts and features.

Referring now to FIG. 1, wherein the present invention will be described with reference to particular embodiments, without limitation, an exemplary embodiment of a vehicle 10 including a steering system 12 is shown. In various embodiments, the steering system 12 includes a steering wheel 14 coupled to a steering shaft 16. In one exemplary embodiment, the steering system 12 is an Electric Power Steering (EPS) system that further includes a steering assist unit 18, the steering assist unit 18 being coupled to the steering shaft 16 of the steering system 12 and to the

tie rods

20, 22 of the vehicle 10. The steering assist unit 18 includes, for example, a rack and pinion steering mechanism (not shown), which may be coupled to a steering actuator motor and a gear transmission (hereinafter referred to as a steering actuator) via the steering shaft 16. During operation, when the steering wheel 14 is turned by a vehicle operator (i.e., driver), the motor of the steering assist unit 18 provides assistance to move the

tie rods

20, 22, which in turn move the

knuckles

24, 26, respectively, the

knuckles

24, 26 being coupled to

road wheels

28, 30, respectively, of the vehicle 10. Although an EPS system is shown in FIG. 1 and described herein, it should be understood that the steering system 12 of the present disclosure may include a variety of controlled steering systems, including but not limited to steering systems having hydraulic configurations, as well as steering via wire configurations.

As shown in FIG. 1, the vehicle 10 also includes various sensors 31-33 that detect and measure observable conditions of the steering system 12 and/or the vehicle 10. The sensors 31-33 periodically or continuously generate sensor signals based on the observable conditions. In various embodiments, sensors 31-33 include, for example, a steering wheel torque sensor, a steering wheel angle sensor, a steering wheel speed sensor, a road wheel speed sensor, and other sensors. In one embodiment, some of these sensors have redundant or backup sensors for validating or supplementing the sensor signals. The sensors 31-33 send signals to the control module 40.

In various embodiments, the control module 40 controls operation of the steering system 12 and/or the vehicle 10 based on one or more enabled sensor signals and also based on the disclosed assist torque calculation systems and methods. Generally, the methods and systems in various embodiments of the present invention generate an assist torque command when a steering wheel torque sensor providing a steering wheel torque signal becomes inactive or fails, without using a steering wheel torque signal that is generally indicative of driver requested assistance. Specifically, the method and system utilize a modified static tire model to estimate rack load or steering rack force when the vehicle is stationary or moving at a relatively low speed (e.g., about 10 kilometers per hour or less). The method and system generate a scaling factor based on a steering wheel angle, a steering wheel speed, a vehicle speed, and a previously generated assist torque command. The method and system generate an assist torque command by scaling the estimated steering rack force by a scaling factor.

FIG. 2 depicts the data flow diagram of FIG. 1 for controlling the steering system 12 and/or the control module 40 of the vehicle 10 of FIG. 1. In various embodiments, the control module 40 may include one or more sub-modules and data storage, such as the rack load estimator 202 and the auxiliary torque command generator 204. As used herein, the terms module and sub-module refer to an Application Specific Integrated Circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or group) and memory that execute one or more software or firmware programs, a combinational logic circuit, and/or other suitable components that provide the described functionality. It will be appreciated that the sub-modules shown in FIG. 2 may be combined and/or further partitioned to similarly generate the assist torque command. It will be appreciated that the sub-modules shown in FIG. 2 may be implemented as a single control module 40 (as shown) or multiple control modules (not shown). Inputs to the control module 40 may be generated from sensors of the vehicle 10 (fig. 1), may be modeled within the control module 40 (e.g., by other sub-modules (not shown)), may be received from other control modules (not shown), and/or may be predefined.

It is well known that ground surfaces contacted by one or more tires of a vehicle and the tires cause rack loads or steering rack forces when the tire plane is rotated (by steering the steering wheel) relative to the surface. In order to steer the steering wheel to the desired position, the steering rack force must be overcome by the torque, in addition to the torque turning the steering wheel. The rack load estimator 202 is configured to estimate a steering rack force and generate an estimated steering rack force signal 212 indicative of the steering rack force based on a steering wheel angle or position signal 206, a steering wheel speed signal 208, and a vehicle speed signal 210. The steering wheel angle signal 206, the steering wheel speed signal 208, and the vehicle speed signal 210 indicate a steering wheel angle value, a steering wheel speed value, and a vehicle speed value, respectively, detected by the various sensors 31-33 of FIG. 1. In some embodiments, rather than generating the steering wheel speed signal 208 by a steering wheel speed sensor, the steering wheel speed signal 208 may be derived from the steering wheel angle signal 206 based on an algorithm that calculates a steering wheel speed value from the steering wheel angle values at different points in time. In some embodiments, the rack load estimator 202 utilizes a modified static tire model to estimate steering rack force. Further details regarding the rack load estimator 202 and the modified static tire model will be described further below with reference to fig. 3.

The assist torque command generator 204 generates an assist torque command 214, which is a periodic or continuous signal indicating an amount of assist torque. The assist torque command 214 is used to command the steering assist unit 18 of fig. 1 to generate an assist torque to assist the driver of the vehicle when the vehicle is stationary or moving at a relatively low speed (e.g., about 10 kilometers per hour (kph) or less). Specifically, the assist torque command generator 204 generates a scaling factor based on the steering wheel angle signal 206, the steering wheel speed signal 208, and the vehicle speed signal 210. The assist torque command generator 204 generates an assist torque command 214 by scaling the estimated steering rack force signal 212 by a scaling factor. Further details regarding the assist torque command generator 204 will be described further below with reference to fig. 4.

In some embodiments, the assist torque command 214 is blended with another assist torque command 216 by a blender 220, the assist torque command 216 also not being generated using the steering wheel torque signal of the steering wheel torque sensor. Specifically, the assist torque command 216 is generated by other sub-modules (not shown) of the control module 40 based on the lateral acceleration of the vehicle estimated from the steering wheel angle signal. In some embodiments, mixer 220 mixes the assist torque commands 214 and 216 by adding commands. Generating the assist torque command 216 is described in U.S. patent application serial No. 14/263,162 filed on 28/4/2014, which is incorporated herein by reference in its entirety. In these embodiments, a mix of assist torque commands 214 and 216 are sent to the motor as assist torque command 218.

Fig. 3 depicts a data flow diagram of the rack load estimator 202 of fig. 2 that estimates steering rack force using a modified static tire model. Static tire models for estimating Steering wheel torque are described in van der Jagt, Pim, "Prediction of Steering force during stationary or Slow Rolling Parking maneuver", Ford der forschunszentrum Aachen GmbH, 10/27 1999, the entire contents of which are incorporated herein by reference. In this disclosure, this static tire model is referred to as the "Van der Jagt static tire model". In some embodiments, the rack load estimator 202 estimates steering wheel torque using a modified Van der Jagt static model.

The Van der Jagt static model includes the following equation for estimating steering rack force caused by a tire and the ground surface contacted by the tire:

M_z＝K_ΨΨ (equation 1)

Wherein K_ΨIs the torsional stiffness of the tire; Ψ is the yaw angle of the wheel plane of the tire; and M_zIs the steering rack force caused by the tire. Different tires have different torsional stiffness.

The Van der Jagt static model also includes the following two equations:

if it is not

If it is not

Wherein

Is the time derivative of the yaw angle Ψ of the wheel plane; Ψ_defIs the torsional deflection (i.e., the deformation angle) of the tire as the steering wheel rotates;

is Ψ_defThe time derivative of (a); m_zmaxIs the maximum torque generated by the tire; and sign () is a function that returns the sign (e.g., positive and negative) of the input value. Equation 2 defines when Ψ_defIs of the same sign as the time derivative of the yaw angle Ψ (i.e., when the direction of deflection of the tire and the direction of the yaw rate of the wheel plane are the same) the torsional deflection Ψ of the tire_defTime derivative of (1)

Equation 3 defines when Ψ_defIs of the same sign as the time derivative of the yaw angle Ψ (i.e., when the direction of deflection of the tire is opposite to the direction of yaw rate at the wheel plane)_defTime derivative of (1)

Equations 2 and 3 illustrate the non-linearity between steering rack force and steering wheel angle.

The Van der Jagt static model also includes the following equation for estimating steering rack force when the vehicle is stationary:

Ψ_defm＝M_zmax/K_Ψ(equation 4)

M_z＝K_Ψ·Ψ_def(equation 6)

Therein Ψ_defmIs the maximum possible deflection of the tire. Equation 4 shows that the maximum possible deflection of the tire before it starts to slip can be calculated by dividing the maximum torque that the tire can generate by the torsional stiffness of the tire. Equation 5 shows the deflection build-up of the tire as the steering wheel rotates. Equation 6 shows that the steering rack force M is estimated by multiplying the torsional stiffness of the tire by the torsional deflection of the tire_z。

The Van der Jagt static model also includes the following equation for estimating steering rack force when the vehicle is moving at a relatively slow speed (e.g., 10kph or less):

τ＝X_rel/(. omega. r) (Eq.8)

Wherein τ is a time constant;

is Ψ_defThe time derivative of (a); x_relIs the tire slack length; ω is the tire rotational speed; and r is the rolling radius of the tire. In the Van der Jagt model, the tire is assumed to have approximately two-thirds of the steady state value as it rolls through the slack length of the tire (e.g., the torsional stiffness and torsional deflection of the tire when the vehicle is stationary). Thus, τ indicates that at time τ, the tire has about two-thirds of its steady state value.

In some embodiments, the rack load estimator 202 includes one or more sub-modules and data storage, such as low pass filters 304 and 306, a maximum torque regulator 308, and the estimation module 302. The rack load estimator 202 estimates steering rack force using a modified Van der Jagt static model. Specifically, the low pass filters 304 and 306 filter the steering wheel angle signal 206 and the steering wheel speed signal 208, respectively. The low pass filters 304 and 306 remove noise from the steering wheel angle signal 206 and the steering wheel speed signal 208 and add time delays to the steering wheel angle signal 206 and the steering wheel speed signal 208. This time delay allows for a more accurate estimation of the steering rack load because the time delay synchronizes the phase of the steering wheel angle signal 206 and the steering wheel speed signal 208 with the motion of the tire. Because the movement of the tire is caused by the movement of the steering wheel, the movement of the steering wheel precedes the movement of the tire.

The estimation module 302 modifies the Van der Jagt static tire model by replacing the tire steering coordinates in equations 1-9 of the Van der Jagt static tire model with the steering wheel angle value, the steering wheel speed value, and the vehicle speed value. For example, the steering wheel angle is used in place of the yaw angle Ψ of the wheel plane of the tire, and the steering wheel speed is used in place of the time derivative of the yaw angle Ψ of the wheel plane

The maximum torque adjuster 308 further modifies the equation of the Van der jagt static tire model by adjusting the maximum torque value that the tire can generate. In the Van der Jagt static tire model, the ground surface is assumed to be a dry road surface. That is, the surface friction is assumed to be constant. To estimate steering rack force from road friction variations, non-linearities, and other unmodeled dynamics, the maximum torque adjuster 308 reduces the maximum torque M that the tire can generate_zmax。

In some embodiments, maximum torque adjuster 308 generates a scaling factor based on steering wheel speed and by dividing M_zmaxMultiplying by a scaling factor to reduce M_zmax. Specifically, the maximum torque regulator 308 uses empirically determinedA threshold steering wheel speed value. The threshold steering wheel speed is used to determine whether the steering wheel speed indicates that the vehicle is on a low friction surface. That is, in some embodiments, if the steering wheel speed is greater than the threshold steering wheel speed, the maximum torque adjuster 308 determines that the vehicle is on a low friction surface (e.g., on icy roads) and sets the scaling factor to a small value (e.g., 1/20 or 0.05). If the steering wheel speed is less than or equal to the threshold steering wheel speed, the maximum torque adjuster 308 determines that the vehicle is not on a low friction surface and sets the scaling factor to a value (e.g., 1) so as not to scale down M_zmax. In some embodiments, the maximum torque adjuster 308 limits the rate at which the scaling factor changes to smoothly scale M_zmax. For example, the maximum torque regulator 308 limits the ramp rate to 0.05 (i.e., the scaling factor is increased such that M is_zmaxRises by a factor of 0.05 times per unit time) and limits the rate of fall to-50 (i.e., the scale factor does not fall more than 50 times per unit time). The maximum torque regulator 308 will M_zmaxMultiplying by a scaling factor to scale M_zmax. The maximum torque adjuster 308 scales the scaled M _zmax310 are sent to the estimation module 302, and the estimation module 302 generates an estimated steering rack force signal 212.

Fig. 4 depicts a data flow diagram of the assist torque command generator 204 of fig. 2. In some embodiments, the assistance torque command generator 204 includes one or more sub-modules and data storage, such as a steering wheel speed based scaling module 402, a steering wheel angle based scaling module 404, a steering wheel speed and angle based limiter 406, a vehicle speed based scaling module 408, a limiter 410, a delay module 412, and

multipliers

414 and 416.

The steering wheel speed based scaling module 402 takes as input the assist torque command 214 and the steering wheel speed signal 208 previously generated by the assist torque command generator 204. The steering wheel speed based scaling module 402 generates a scaling factor 420 that is used to scale down the estimated steering rack force signal 212. The estimated steering rack force signal 212 is scaled by a scaling factor 420 such that the output assist torque command 214 generated from the estimated steering rack force signal 212 achieves a natural return of the steering wheel to the center position without the driver providing torque to the steering wheel.

In some embodiments, the steering wheel speed-based scaling module 402 sets the scaling factor 420 to a value (e.g., 0.3) to ramp down the estimated steering rack force signal 212 to 30% when the steering wheel speed is less than the threshold speed. When the steering wheel speed is greater than the threshold speed, the steering wheel speed-based scaling module 402 sets a scaling factor 420 to ramp the estimated steering rack force signal 212 to a full value (e.g., approximately 100%). The scaling factor 420 is used to ramp up the estimated steering rack force signal 212 when the assist torque command 214 indicates assist torque in the same direction as the steering wheel speed signal 208. The scaling factor 420 is used to ramp down the assist torque command when the assist torque command is in the opposite direction of the steering wheel speed (i.e., when the assist torque command 214 and the steering wheel speed have different signs-quadrants II and IV). An example of a steering wheel speed based scaling module 402 is described in the above-incorporated U.S. patent application serial No. 14/263,162.

The steering wheel angle based scaling module 404 takes as input the assist torque command 214, the vehicle speed signal 210, and the steering wheel angle signal 206, which were previously generated by the assist torque command generator 204. The steering wheel angle based scaling module 404 generates a scaling factor 422 for scaling down the estimated steering rack force signal 212. The estimated steering rack force signal 212 is scaled by a scaling factor 422 such that the output assist torque command 214 generated from the estimated steering rack force signal 212 achieves a natural return of the steering wheel to center without the driver providing torque to the steering wheel. Further details of the steering wheel angle based scaling module 404 are described below with reference to fig. 5.

The steering wheel speed and angle based limiter 406 takes as input the steering wheel speed signal 208 and the steering wheel angle signal 206. The steering wheel speed and angle based limiter 406 generates a scaling factor 424 for scaling down the estimated steering rack force signal 212. The estimated steering rack force signal 212 is scaled by a scaling factor 424 such that the output assist torque command 214 generated from the estimated steering rack force signal 212 does not overly assist the driver (i.e., does not provide more assist torque than is required).

In some embodiments, the steering wheel speed and angle based limiter 406 determines the first gain value using a first gain table indexed by the steering wheel angle value indicated by the steering wheel angle signal 206. The first gain table returns a constant gain (e.g., 1) for steering wheel angle values below the threshold steering wheel angle. As the steering wheel angle value increases, the gain value returned by the first gain table becomes smaller for steering wheel angle values above the threshold steering wheel angle. Similarly, steering wheel speed and angle based limiter 406 uses a second gain table indexed by the steering wheel speed value indicated by steering wheel speed signal 208 to determine a second gain value. The second gain table returns a constant gain (e.g., 1) for steering wheel speed values below the threshold steering wheel speed. As the steering wheel speed value increases, the gain value returned by the second gain table becomes smaller for steering wheel speed values above the threshold steering wheel speed. The limiter 406 based on the steering wheel speed and angle multiplies the first gain value by the second gain value. Then, the steering wheel speed and angle-based limiter 406 limits the rate of change of the product of the first gain value and the second gain value within a range such that the value of the product changes smoothly. The resulting product is the scaling factor 424.

The vehicle speed based scaling module 408 takes as input the vehicle speed signal 210. The vehicle speed based scaling module 408 generates a scaling factor 426 that is used to scale down the estimated steering rack force signal 212. The estimated steering rack force signal 212 is scaled by a scaling factor 426 such that the output assist torque command 214 generated from the estimated steering rack force signal 212 gradually shrinks to zero as the vehicle speed increases. Specifically, in some embodiments, the vehicle speed based scaling module 408 determines the speed related gain using a speed related gain table indexed by the vehicle speed value indicated by the vehicle speed signal 210. The gain value returned by the speed-dependent gain table becomes larger as the vehicle speed increases. Once the vehicle speed reaches above the threshold vehicle speed, the gain value saturates. The vehicle speed based scaling module 408 limits the gain value to a range (e.g., a range from 0 to 1). The resulting gain value is the scaling factor 426.

In some embodiments, multiplier 414 multiplies together the four scaling

factors

420, 422, 424, and 426 and sends the product of the four scaling factors to limiter 410, which limiter 410 limits the product to a range (e.g., 0 to 1). The multiplier 416 then generates the output assist torque command 214 by multiplying the estimated steering rack force by the product of the four scaling factors. The output assist torque command 214 is delayed, e.g., by a unit time, by a delay module 412 and then provided to a steering wheel speed based scaling module 402 and a steering wheel angle based scaling module 404. Additionally, as described above with reference to fig. 2, in some embodiments, the assist torque command 214 is blended with the assist torque command 216.

Fig. 5 depicts a data flow diagram of the steering wheel angle based scaling module 404 of fig. 4. In some embodiments, the steering wheel angle based scaling module 404 includes one or more sub-modules and data storage, such as a gain determiner 502, a vehicle speed dependent gain table 504, a limiter 506, a subtractor 508, sign

determiners

510 and 512, a multiplier 514, a selector 516, a multiplier 518, a mixer 520, a limiter 522, and a rate limiter 524. As described above, the steering wheel angle based scaling module 404 takes as input the assist torque command 214 previously generated by the assist torque command generator 204, the vehicle speed signal 210, and the steering wheel angle signal 206.

The gain determiner 502 determines a speed dependent gain signal 526 based on the vehicle speed 210. Specifically, in some embodiments, the gain determiner 502 uses a vehicle speed dependent gain table 504 indexed by the vehicle speed value indicated by the vehicle speed signal 210. The speed dependent gain table 504 is a vehicle speed return constant (e.g., 1) below a threshold vehicle speed. As the vehicle speed value increases, the gain value returned by the speed dependent gain table 504 for vehicle speed values above the threshold vehicle speed becomes smaller.

The limiter 506 limits the speed-dependent gain signal 526 to a range of gain values (e.g., a range from 0 to 1) to generate a limited speed-dependent gain signal 528. The subtractor 508 then subtracts the limited velocity-dependent gain signal 528 from a constant 530 (e.g., 1) to generate a gain signal 532.

Symbol determiners

510 and 512 each receive an input signal and generate a symbol signal based on the sign of the input signal value. For example, when the input signal indicates a negative value, the sign determiner generates-1. The sign determiner generates +1 when the input signal indicates a positive value. The sign determiner generates a zero when the input signal indicates a zero. Sign determiner 510 takes assist torque command 214 as an input signal and generates sign signal 534. The sign determiner 512 takes the steering wheel angle signal 206 as an input signal and generates a sign signal 536.

The multiplier 514 generates a quadrant signal 538 by multiplying the two

sign signals

534 and 536. When the quadrant signal 538 indicates a negative value, this means that the sign of the assist torque command 214 is different from the sign of the steering wheel angle 215 (i.e., the steering wheel angle value and the assist torque value constitute the second quadrant or the fourth quadrant in the two-dimensional coordinate system of the two axes). That is, the steering wheel is steered to the left of the center position with the assist torque indicated by the assist torque command 214 directed to the right, or the steering wheel is steered to the right of the center position with the assist torque directed to the left. When the quadrant signal 538 indicates a positive value, it means that the sign of the assist torque command 214 is the same as the sign of the steering wheel angle 215 (i.e., the first quadrant or the third quadrant). That is, the steering wheel is steered to the left of the center position with the assist torque indicated by the assist torque command 214 directed to the left, or the steering wheel is steered to the right of the center position with the assist torque directed to the right. When the quadrant signal 538 is zero, it means that the steering wheel is in a center position, or the assist torque indicated by the assist torque command 214 is zero (i.e., the steering wheel is stationary).

Based on the quadrant signal 538, the selector 516 generates a gain signal 540. Specifically, if the quadrant signal 538 indicates a negative value, the selector 516 selects the quadrant-based gain value 544 as the gain signal 540. In some embodiments, the quadrant-based gain values 544 are predetermined based on different possible quadrant signal values. If the quadrant signal 538 does not indicate a negative value (i.e., the quadrant signal 538 indicates a positive value or zero), the selector 516 selects a constant 542 (e.g., 1) as the gain signal 540.

Multiplier 518 multiplies gain signal 532 from subtractor 508 with gain signal 540 from selector 516 to generate scaling factor 546. The mixer 520 mixes (e.g., adds) the scaling factor 546 to the limited speed-based gain signal 528 from the limiter 506 to generate the scaling factor 548. The limiter 522 limits the scaling factor 548 to a range of gain values (e.g., a range from 0 to 1) to generate a limited speed factor 550. The rate limiter 524 then limits the rate of change of the limited scaling factor 550 to a range such that the value of the limited scaling factor 550 changes smoothly over time. The output signal of rate limiter 524 is scaling factor 422.

Referring now to FIG. 6, a flowchart illustrates an assist torque command generation method that may be performed by the control module 40 of FIG. 1. It will be understood from this disclosure that the order of operations in the method is not limited to being performed in the order shown in fig. 6, but may be performed in one or more varying orders as applicable and in accordance with this disclosure. In various embodiments, the method may be scheduled to run based on a predetermined event, and/or run continuously during operation of the vehicle 10.

At block 610, the control module 40 receives sensor signals from the sensors 31-33 of FIG. 1. The control module 40 then determines whether one or more steering wheel torque sensors of the vehicle 10 are enabled or functioning properly at block 620. The control module 40 may determine whether the steering wheel torque sensor is enabled by, for example, analyzing the steering wheel torque signal from the sensor. When the control module 40 determines that one or more steering wheel torque sensors are not enabled, the control module 40 proceeds to block 640, which is described further below. When the control module 40 determines that one or more steering wheel torque sensors are enabled and that at least one steering wheel torque sensor signal is available, the control module 40 generates an assist torque command using the torque sensor signal at block 630.

At block 640, the control module 40 estimates or predicts a steering rack force caused by the tires of the vehicle and the ground surface contacted by the tires when the vehicle is stationary or moving at a relatively low speed that is below a threshold speed. In some embodiments, the control module 40 estimates steering rack force using a modified static tire model. The control module 40 may filter the steering wheel angle signal 206 and the steering wheel speed signal 208 using low pass filters 304 and 306, respectively, to remove noise from these signals and apply a delay to the signals. The control module 40 may also narrow the maximum amount of torque that the tire can generate based on the vehicle speed signal 210.

At block 650, the control module 40 generates an assist torque command 214 based on the steering rack force estimated at block 640. Specifically, in some embodiments, the control module 40 uses a product of a plurality of scaling factors to scale down the estimated steering rack force to generate the assist torque command 214 based on the estimated steering rack force. The control module 40 generates a scaling factor based on the previously generated assist torque command 214, the vehicle speed signal 210, and the steering wheel angle signal 206. The control module 40 generates another scaling factor based on the steering wheel angle signal 206 and the steering wheel speed signal 208. The control module 40 generates another scaling factor based on the assist torque command 214, the vehicle speed signal 210, and the steering wheel angle signal 206. The control module 40 generates another scaling factor based on the vehicle speed signal 210.

At block 660, the control module 40 optionally blends the assist torque command generated at block 640 with assist torque commands that the control module 40 may generate. In some embodiments, the control module 40 generates another assist torque command 216 based on the lateral acceleration of the vehicle estimated from the steering wheel angle signal.

At block 670, the control module 40 controls the EPS system by sending the assist torque command generated at

block

630 or 650 or the hybrid value generated at block 660 to the motor of the EPS system.

Fig. 7 depicts a data flow diagram of the rack load estimator 202 of fig. 2 utilizing another example of a modified static tire model and an example linear spring model to estimate steering rack force. In the illustrated example, the estimation module 302 uses the vehicle speed 210, the motor angle 704, the motor speed 706, and the coefficient of friction 710 to generate the estimated steering rack force 212. The motor angle 704 and motor speed 706 are obtained from the actuator motor in the steering assist module 18. Alternatively or additionally, the estimation module 302 uses the steering wheel angle 206 and the steering wheel speed 208 as inputs. The motor angle 704 may pass through the low pass filter 306 and the motor speed 706 may pass through the low pass filter 304 before estimation. The friction estimator module 708 calculates a friction coefficient 710.

Estimated steering rack force 212M output by the estimation module 202 of FIG. 7_zCan be specified as

M_z＝M_z1+M_z2Wherein (equation 10)

M_z1＝μ·K_Ψ·Ψ_defAnd (equation 11)

M_z2＝K_Ψ2Ψ (equation 12)

In the above equation, M_zIs represented by M_z1And M_z2Estimated alignment torque of composition, where M_z1Representing the torque component due to tire stiffness (see equation 6 above), M_z2Representing the torque component caused by the linear spring. As described above, Ψ represents the tire angle, and K_ΨRepresenting the torsional tire stiffness at angle Ψ. Furthermore, K_Ψ2Represents the linear spring rate at the angle Ψ, and μ represents the coefficient of friction 710 estimated by the friction estimator 708. The linear spring rate or the torsional tire rate represents the torque caused by the slip experienced by the vehicle tires, particularly at low vehicle speeds. Additionally or alternatively, the linear spring rate represents a torque caused by a particular geometry of a suspension of the vehicle.

Further, the estimation module 202 of fig. 7 may calculate (Ψ) as the motor rotates, as compared to equations 2 and 3_def) Time derivative of torsional deflection of a tire

Is composed of

If it is not

(equation 13) and

if it is not

In addition, in contrast to equation 5 above, in the estimation module of fig. 7,

wherein (equation 15)

Ψ_def＝sign(Ψ_def,0).min(Ψ_def,max,|Ψ_def,0L), wherein (equation 16)

Ψ_def,max＝μ.M_z1,max/K_ΨAnd wherein (equation 17)

Wherein τ ═ X_rel/(ω, r) (equation 18)

Thus, the example estimation module 202 of fig. 7 uses the coefficient of friction 710(μ) to estimate the steering rack force 212 (M)_z). In one or more examples, friction estimator 708 calculates friction coefficient 710 at low vehicle speeds (e.g., 0-20 kph).

FIG. 8 illustrates example components and example data flows of the friction estimator 708. The friction estimator 708 includes hardware such as electronic circuitry. Additionally, the friction estimator 708 may include computer-executable instructions that control the operation of the friction estimator 708. Additionally, the friction estimator 708 may include a memory that stores predetermined data and/or data generated and used during operation of one or more components of the friction estimator 708. For example, the friction estimator 708 includes components such as a speed-to-fast friction converter 810, a friction learning module 820, a rate limiter module 830, and a friction saturation module 840, among other components.

For example, the friction estimator 708 receives the motor speed (u) as an input 802. In one or more examples, the friction estimator 708 calculates an absolute value of the motor speed in further operation, as shown in block 805. The motor speed is forwarded to a speed-to-fast friction converter 810. The speed-to-rapid friction converter 810 converts the motor speed value to a rapid friction value μ_fast. In one or more examples, the speed-to-fast friction converter 810 uses a one-dimensional lookup table to generate μ_fastThe value is obtained. For example, the lookup table includes μ for known static motor speed values_fastThe signal value. Mu corresponding to the current motor speed_fastThe values are forwarded to the friction learning module 820 and the rate limiter 830. In one or more examples, μ is determined and continuously forwarded_fastValue, hence μ_fastThe value may be referred to as μ_fastA signal.

Rate limiter 830 limits μ_fastThe rate of change (or bandwidth) of the signal. For example, the rate limiter 830 may correspond to μ at a lower (i.e., limited) rate of change (or bandwidth) than the frequency at which the motor speed is input to the friction estimator 708_fast-limitedThe signal is output to the friction saturation module 840. Thus, the rate limiter 830 limits the rate of change (derivative) of the output signal, i.e., if the input of the rate limiter changes from 0 to 1 in one second, the output may change from 0 to 1 in more than one second. Thus, the rate limiter 830 is used to limit the rate of change (up or down) of the output. By limiting the rate of change (or bandwidth) in this manner, rate limiter 830 helps to change the coefficient of friction 710 after a predetermined period of time and avoids abrupt changes in the friction estimate. Sudden changes in the friction estimate may change the estimated steering rack force 212, which may result in undesirable changes in the static/low speed assist 218.

Friction learning module 820 analyzes μ_fastSignal and will correspond to mu_slowThe signal is output to the friction saturation module 840. Friction learning module 820 outputs μ at a specific ratio_fastFrequency (or rate) calculation mu of signal with slow frequency_slowA signal. Furthermore, μ is updated only under certain conditions_slowSignals, as described later. The friction saturation module 840 receives μ limited by a variable upper limit and a predetermined lower limit_fast-limitedA signal. Mu.s_slowThe signal is determined to be applied to μ by the friction saturation module 840_fast-limitedThe upper limit of the signal. The friction saturation module 840 also accesses a constant (C) that is a predetermined lower threshold level (limit) for the estimated friction coefficient 710, as shown in block 835. The constant C may have different values in different examples, e.g. 0, 0.01, 0.5, etc. Friction saturation Module 840 bases μ from friction learning Module 820_slowSignal, μ limited by rate limiter 830_fast-limitedThe signal and a constant C to output an estimated coefficient of friction 710(μ). Thus, the friction estimator 708 uses the motor speed to predict the friction coefficient 710(μ).

FIG. 9 illustrates exemplary components and exemplary data flows of the friction learning module 820. In one or more examples, friction learning module 820 includes rate limiter 910, converter 920, and flag updater 930, among other components. As described above, the friction learning module 820 receives μ_fastThe signal is used as input. Rate limiter 910 to match the input mu_fastReceiving input mu at different frequencies_fastSignal and output the second mu_fast2A signal. Converter 920 receives mu_fast2The signal is taken as input and the corresponding mu is generated_slowA signal. Converter 920 also receives μ from flag updater 930_update-flagAs an input.

The flag updater 930 bases on μ_slowSum of signals mu_fast2Previous value of signal, will_update-flagIs set to TRUE (TRUE, 1) or FALSE (FALSE, 0). FIG. 10 depicts example components and example data flows of flag updater 930. It should be understood that the flag updater of FIG. 9 is an example, and that other implementations of flag updater 930 are possible. If μ_fast2Value of signal ≤ mu_slowPrevious value of signal (i.e. output of 1050) and if μ_fast2If the signal is decreasing, then the flag updater 930 shown in FIG. 10 will cause μ to be asserted_update-flagThe value of (d) is set to 1. For example, as shown in the figure,the tag updater 930 receives μ_fast2Sum of signals mu_slowSignal as input and output mu_update-flagAs shown in

blocks

1010, 1020, and 1030.

Received mu_fast2The signal is passed through a unit delay to provide mu_fast2The previous value of the signal, as indicated in block 1040. At the same time, the received mu_fast2The signal is used for determining mu through a subtracter_fast2Whether the signal is decreasing, as indicated at block 1060. By passing from mu_fast2Subtracting mu from the previous value of the signal (delayed signal)_fast2Determining mu from the current value of the signal_fast2Whether the signal is decreasing. The result of the subtractor is compared to a constant (0; zero) to determine the sign (positive or negative) of the result, as shown in block 1070. If the output of the sign comparison is positive, then the previous value is greater than the current value, μ_fast2The signal is decreasing (and vice versa). Therefore, if μ_fast2The signal is decreasing, the output of the sign comparison is 1, TRUE, otherwise 0 (FALSE). The output of the sign comparison is provided to a logical operator, such as an AND gate, as shown in block 1080. In addition, the logical operators receive input from relational operators, which inputs will be μ_fast2Value of signal and mu_slowThe previous values of the signals are compared as indicated in block 1080.

E.g. received mu_slowThe signal is passed through a unit delay to provide mu_slowThe previous value of the signal, as indicated in block 1050. The relational operator receives mu_slowPrevious value of signal and mu_fast2The current value of the signal (i.e. the received mu)_fast2Signal) as shown in block 1090. The relational operator compares two input values and if μ_fast2Signal ≤ mu_slowA signal value, 1 (TRUE) is output, otherwise 0 (FALSE) (or vice versa). Thus, the logical operator receives two inputs- μ_fast2Whether the signal is less than or equal to mu_slowAn indication of a previous value of the signal value; and mu_fast2An indication of whether the signal is decreasing, as shown in block 1080. If both of these conditions are true, then flag updater 930 will μ_update-flagIs set to 1 (TRUE),otherwise, it is set to 0 (FALSE) (or vice versa).

Referring back to FIG. 9, converter 920 is based on μ_update-flagMu to_fast2Conversion of signal into mu_slowA signal. For example, if μ_update-flagIs 0 (zero; false), converter 920 does not perform the conversion and remains as μ_update-flagIs the most recent output at TRUE, and if μ_update-flagIs 1 (true), converter 920 performs the conversion (or vice versa). FIG. 11 shows μ being converted by converter 920_fast2Conversion of signal into mu_slowExample components of a signal and example data flows. It should be understood that the converter 920 depicted in fig. 11 is exemplary, and other implementations may include other converters.

The example converter 920 of FIG. 11 uses filters and attenuations to convert μ_fast2Conversion of signal into mu_slowA signal. Converter 920 receiving mu_fast2After the signal, μ is subtracted from the constant C1_fast2Signal, as shown in block 1102. The constant value may be, for example, μ_fast2A maximum threshold value for the signal, such as 1, or any other predetermined constant value. As shown in block 1110, the saturation block of converter 920 receives the subtracted output. The saturation block limits the magnitude of the subtraction output based on a predetermined upper limit and a predetermined lower limit. For example, the saturation block compares the subtraction output to an upper limit and a lower limit. If the subtraction output is less than the lower limit, the saturation block outputs the lower limit. Similarly, if the subtraction output is greater than the upper limit, the saturation block outputs the upper limit. In other cases, the saturation block outputs the subtraction output.

The output from the saturation block is passed to a selector, as shown in block 1120. Output of selector at saturation block and based on mu_update-flagIs switched between delayed outputs. As shown in block 1120, the selector receives μ_update-flagAnd a delayed output as other inputs. The delayed output is received by passing the output of the filter module of converter 920 through a unit delay, as shown in

blocks

1130 and 1140. The filter module may be a low pass filter that receives the output of the selector as one input and a predetermined low pass frequency as a second input, as indicated by block 1140Shown in the figure. Alternatively or additionally, the low pass filter module generates the output according to:

y [ n ] ═ b1.u [ n ] + b2.u [ n-1] -k.a 2.y [ n-1] (equation 19)

Where y is the output of the filter and u is the input to the filter, in some embodiments, predetermined low pass frequency values are used to determine coefficients (b1, b2, and a2) for a simple first order low pass filter, and K may be a value between 0.99 and 1.0. These parameters facilitate the filter module output y to decay slowly with respect to time to zero when K is less than 1. When K is 1, y (output) remains unchanged, assuming the input is constant.

As depicted, converter module 920 receives μ_update-flagAs an input, based on the input, the module output is updated to a new μ_slowThe value, or the previous value itself (i.e., the output is not updated). Therefore, if μ_update-flagIf true, converter 920 processes μ through a digital filter_fast2Signal to generate new mu_slowOutput the value, and if μ_update-flagIf false, the converter 920 holds the output. The function of converter 920 can be described by the following example digital filter equation. In these equations, by subtracting μ from 1_fast2The signal gets the input of a digital filter u. In addition, the output y of the filter is subtracted from 1 to obtain the output μ_slow。

y[n]＝b1.u[n]+b2.u[n-1]–K.a2.y[n–1]If μ_update-flagTrue; and (Eq. 20)

y[n]＝y[n–1]If μ_update-flagFalse (equation 21)

Where y is the output of the filter, u is the input of the filter, n indicates the current calculation step, the coefficients (b1, b2, and a2) are predetermined parameter values for a simple first order low pass filter based on predetermined low pass frequency values, and K is a value less than or equal to 1.

Thus, the estimation module 302 of fig. 7 predicts the static torque generated by the tire by modifying the Van der Jagt static model using the spring rate model to provide additional torque. Additionally, the estimation module 302 of fig. 7 estimates the coefficient of friction 710 to calculate the estimated steering rack force 212. The modified Van der Jagt model is represented by equations 10-18.

Fig. 12 shows an alternative example of a mixer 220 that mixes two or more assist torque commands. The assist torque command generator 204 generates the assist torque command 214 for static or low vehicle speeds (0-20kph) based on the estimation module 302, which the estimation module 302 operates according to the modified Van der Jagt model as described herein. The assist torque command generator 204 may also generate an assist torque command 216, which assist torque command 216 may be an assist torque command for a higher speed (e.g., >20kph) vehicle. An example of generating such an assist torque command at higher speeds is described in U.S. patent application serial No. 14/263,162 filed on 28/4/2014, which is incorporated herein by reference in its entirety. The mixer 220 mixes the assist torque commands 214 and 216 and sends a mixing command 218 to the motor.

In the exemplary mixer 220 of fig. 12, mixing is based on vehicle speed. For example, the mixer 220 receives a current vehicle speed as an input and determines a scaling factor S based on the vehicle speed_xAs shown in block 1210. For example, the mixer 220 includes a one-dimensional look-up table that includes a plurality of mappings between vehicle speeds and scaling factors. Alternatively, or additionally, the scaling factor may be increased according to a predetermined linear relationship of vehicle speed. In one or more examples, the scaling factor may be increased until the vehicle speed reaches a predetermined threshold, after which the scaling factor may be constant. In one or more examples, mixer 220 may include a saturation block that limits the scaling factor to within a predetermined range, as shown at block 1220. For example, the predetermined range may be 0 to 1. In this case, the scaling factor may be normalized to a predetermined range of saturated blocks.

Mixer 220 uses the scaling factor and scroll assist 216 as inputs to the multiplier, as shown in block 1230. The multiplier outputs the product of the scaling factor and the scroll assist 218. The mixer 220 will predetermine the constant C₀Adding to the scaling factor S_xAnd the result (-S) is used_x+C₀) As multipliersThe first input, the multiplier also receives the static/low speed assist 214 as a second input, as shown in

blocks

1240 and 1250. Mixer 220 further sums the outputs of the two products from scroll assist 216 and static/low speed assist 214 to provide mix assist instructions 218, as shown in block 1260. Thus, an EPS implementing the solution described herein facilitates mixing two or more torque assist commands that may be generated at different vehicle speeds depending on speed in an improved manner.

The solution described herein facilitates learning the coefficient of friction applied to the tire stiffness model when torque assistance is provided by the EPS system, particularly when the vehicle is at static or low vehicle speeds (0-20 kph). Based on the coefficient of friction, the solution described herein maintains a reduction in tire model force when a low coefficient of friction is detected. Furthermore, the solution uses the friction coefficient to limit the value of the calculated tyre wind-up (e.g. low friction reduces the maximum possible wind-up). By utilizing the above determination and use of the coefficient of friction, the EPS assist system avoids low frequency oscillations due to speed feedback, and also avoids undesirable transitions between high and low friction assist values.

In addition, the solution described herein includes a linear spring contribution in addition to the tire model when determining the estimated steering rack force. The linear spring contribution helps the EPS system to take into account the additional forces affecting the steering wheel due to the suspension geometry of the vehicle. By taking into account the linear spring contribution, the EPS system may provide additional assistance to support additional forces caused by suspension geometry, particularly when the vehicle is stationary or traveling at low speeds (0-20 kph).

Further, the solution described herein facilitates the EPS system to mix multiple aids provided to the steering wheel depending on the vehicle speed. For example, the technical solution facilitates the EPS system to mix a first assistance instruction from a tire model used at low vehicle speeds (0-20kph) with a second assistance instruction from a rolling bicycle model used at higher speeds (>20 kph).

Further, the solutions described herein facilitate the EPS system to provide torque assistance on surfaces with low friction (e.g., ice), which may result in the vehicle indicating vehicle speed as static, i.e., 0kph, even when the vehicle may be rolling (such as at 2-4 kph). For example, the technical solution facilitates the EPS system to exclude assistance at lower speeds (e.g. <4kph) depending on vehicle sensor performance.

While the invention has been described in detail in connection with only a limited number of embodiments, it should be readily understood that the invention is not limited to such disclosed embodiments. Rather, the invention can be modified to incorporate any number of variations, alterations, substitutions or equivalent arrangements not heretofore described, but which are commensurate with the spirit and scope of the invention. Additionally, while various embodiments of the invention have been described, it is to be understood that aspects of the invention may include only some of the described embodiments. Accordingly, the invention is not to be seen as limited by the foregoing description.

Claims

1. A method of controlling an electric power steering system, the method comprising:

determining a steering rack force to apply a torque to the electric power steering system, wherein determining the steering rack force comprises: generating a coefficient of friction based on a motor speed of a motor of the electric power steering system, the coefficient of friction being indicative of friction between a tire and a ground surface with which the tire is in contact;

generating a steering assist torque command based on the steering rack force; and

controlling the electric power steering system using the steering assist torque command,

wherein generating the coefficient of friction comprises:

converting the motor speed to a first coefficient of friction value;

determining an upper limit for the coefficient of friction based on the first coefficient of friction value; and

scaling the first coefficient of friction value to determine a coefficient of friction value.

2. The method of claim 1, wherein the steering rack force is generated in response to determining that one or more steering wheel torque sensors of the electric power steering system are not enabled.

3. The method of claim 1, wherein determining the steering rack force further comprises: determining a component of the steering rack force caused by the tire and a ground surface contacted by the tire based on the motor speed.

4. The method of claim 3, wherein determining the component of the steering rack force further comprises:

filtering a motor angle using a low pass filter, the motor angle being an angle of a force experienced by the motor of the electric power steering system;

filtering the motor speed using a low pass filter; and

estimating the steering rack force using the filtered motor angle and the filtered motor speed.

5. The method of claim 3, wherein the component is a first component, and determining the steering rack force further comprises: a second component of the steering rack force caused by the linear spring rate is determined.

6. The method of claim 1, wherein the upper limit of the coefficient of friction is determined in response to a value of the first coefficient of friction being less than an immediately preceding value of the motor speed.

7. The method of claim 1, wherein the upper limit of the friction coefficient is determined in response to the first friction coefficient value being less than or equal to a delay value of the upper limit of the friction coefficient.

8. The method of claim 1, wherein the upper limit of the coefficient of friction is determined in response to:

the value of the first coefficient of friction is less than the immediately preceding value of the first coefficient of friction; and

the first coefficient of friction value is less than or equal to the upper limit of the coefficient of friction.

9. A system, comprising:

a power steering system including one or more steering wheel torque sensors; and

a control module configured to:

determining a steering rack force to apply a torque to a steering wheel, wherein determining the steering rack force comprises: generating a coefficient of friction based on a motor speed of a motor of the power steering system, the coefficient of friction being indicative of friction between a tire and a ground surface with which the tire is in contact;

controlling the power steering system using the steering assist torque command,

wherein generating the coefficient of friction comprises:

converting the motor speed to a first coefficient of friction value;

10. A power steering system comprising:

a steering assist unit; and

a control module coupled with the steering assist unit, the control module configured to:

determining a steering rack force to apply a torque to a steering wheel, wherein determining the steering rack force comprises: generating a friction coefficient based on a motor speed of a motor of the steering assist unit, the friction coefficient being indicative of friction between a tire and a ground surface with which the tire is in contact;

controlling the power steering system using the steering assist torque command, wherein generating the coefficient of friction comprises: determining an upper limit for the friction coefficient based on a first friction coefficient value converted from the motor speed in response to the value of the motor speed being less than an immediately preceding value of the motor speed.

11. The power steering system of claim 10, wherein the steering rack force is generated in response to determining that one or more steering wheel torque sensors are not enabled.

12. The power steering system of claim 10, wherein determining the steering rack force further comprises: determining a component of the steering rack force caused by the tire and a ground surface in contact with the tire based on the motor speed.

13. The power steering system of claim 12, wherein the component is a first component, and determining the steering rack force further comprises: a second component of the steering rack force caused by the linear spring rate is determined.

14. The power steering system of claim 10, wherein the steering assist torque command is a first steering assist torque command, and the control module is further configured to:

generating a second steering assist torque command;

generating a third steering assist torque command by combining the first steering assist torque command and the second steering assist torque command; and

controlling the power steering system using the third steering assist torque command.